The long-term objective of this research is the quantification of acute and chronic effects of alcohol on electromotive generators in the brain, and the determination of relationships of abnormal generators and predisposition to excessive alcohol consumption. Research integrates the development of high resolution EEG methods to localize and study electromotive generators of brain processes with various electrophysiological models, both computational and theoretical, to validate the methods and test hypotheses of suspected generators. This integrated procedure is used to determine how, when, and where alcohol alters mechanisms which process sensory input, respond to input, and store and retrieve information. Event-related brain electrical potentials (ERPs), which provide both spatial and temporal information about these mechanisms, and nonstationary EEGs, which provide a measure of the brain's dynamic interactions, are used in this investigation. A physical, artificial head model is also used to study the inhomogeneous and anistropic nature of volume conduction in the head. Methods have been developed that improve the spatial resolution of current EEG methods by a factor of three. Preliminary findings suggest that temporal resolution is improved as well. These methods have been shown to provide better estimates of underlying brain processes than present techniques and are more insensitive to electrophysiological artifacts, such as eye blinks. These new techniques are currently being used to analyze EEG data of individuals with different family histories of alcoholism.